Our proprietary AI engine that analyzes disease progression across multiple visits — something no other system does
Most diagnostic tools look at single snapshots:
"Patient has elevated liver enzymes today" → Treated as isolated event
"Patient has neurological symptoms today" → Misdiagnosed as Parkinson's
Doctor doesn't remember what happened 8 months ago → Pattern invisible
But rare diseases reveal themselves over time. Wilson disease starts with liver issues, then months/years later shows neurological symptoms. If you only look at today's visit, you miss the pattern.
Three dimensions that create a unique "signature" for each disease
How fast is the disease moving?
Acute (days-weeks)
Anti-NMDAR Encephalitis: Psychiatric → seizures in days
Subacute (months)
Wilson Disease: Liver → neuro symptoms over 6-12 months
Chronic (years-decades)
Fabry Disease: 10+ years before major organ damage
Knowing the tempo narrows possibilities from 7,000 rare diseases to a few hundred.
Which organ systems are affected and in what order?
Liver → Brain
Wilson Disease (copper accumulation)
Kidney + Heart + Nerves
Fabry Disease (multi-system involvement)
Brain + Eyes + Muscles
MELAS (mitochondrial disorder)
The pattern of organ involvement is like a fingerprint for each rare disease.
Is the disease progressive, episodic, or stable?
Relentlessly Progressive
Pompe Disease: Steady muscle weakness decline
Episodic (Flare-ups)
Hereditary Angioedema: Sudden swelling attacks
Stepwise Decline
MELAS: Stroke-like episodes with progressive damage
The trajectory tells you about disease mechanism and treatment urgency.
Tempo
Subacute (6-18 months)
Topology
Liver → Brain (in order)
Trajectory
Progressive worsening
When LIET sees this combination, it alerts:
⚠️ TTT Pattern Match: Wilson Disease (copper accumulation disorder)
Recommend: Ceruloplasmin test + genetic screening for ATP7B gene
Uses TTT to analyze real patient visits in real-time, detecting rare disease patterns that emerge across months/years of data.
Uses TTT to visualize health timelines and alert families when patterns match known rare disease signatures.
Uses TTT to generate realistic disease progressions for synthetic patients, ensuring they match real-world patterns.